trainer_state.json

{
  "best_metric": 0.8877333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0351/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 38.93034362792969,
      "learning_rate": 0.00047524221697560476,
      "loss": 1.2972,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8165333333333333,
      "eval_loss": 0.6101328134536743,
      "eval_runtime": 7.7878,
      "eval_samples_per_second": 481.522,
      "eval_steps_per_second": 7.576,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 111.42781066894531,
      "learning_rate": 0.0004058724504646834,
      "loss": 0.4406,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.472905695438385,
      "eval_runtime": 7.6195,
      "eval_samples_per_second": 492.155,
      "eval_steps_per_second": 7.743,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.16506576538086,
      "learning_rate": 0.0003056302334890786,
      "loss": 0.2527,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.48740094900131226,
      "eval_runtime": 7.6111,
      "eval_samples_per_second": 492.702,
      "eval_steps_per_second": 7.752,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 21.602155685424805,
      "learning_rate": 0.00019436976651092142,
      "loss": 0.1321,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.4546962380409241,
      "eval_runtime": 7.6035,
      "eval_samples_per_second": 493.192,
      "eval_steps_per_second": 7.76,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.69365692138672,
      "learning_rate": 9.412754953531663e-05,
      "loss": 0.0711,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.4425203502178192,
      "eval_runtime": 7.7177,
      "eval_samples_per_second": 485.897,
      "eval_steps_per_second": 7.645,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.522737503051758,
      "learning_rate": 2.4757783024395242e-05,
      "loss": 0.038,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.4362659156322479,
      "eval_runtime": 7.5585,
      "eval_samples_per_second": 496.129,
      "eval_steps_per_second": 7.806,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}